<![CDATA[The development of Software-Defined Networking technology offers high flexibility in network management but also introduces security vulnerabilities to Distributed Denial of Service attacks. These attacks can paralyze the SDN controller through packet flooding, leading to a drastic decline in network performance. This research aims to enhance the accuracy of DDoS attack detection in an SDN architecture based on the Ryu Controller with the OpenFlow v1.3 protocol by implementing the Random Forest algorithm optimized with Gain Ratio feature selection. The novelty of this study lies in the use of a primary dataset collected through a self-developed SDN testbed using Mininet with a Modified Tree topology to generate more realistic attack scenarios. The Gain Ratio method is employed to reduce data dimensionality by selecting the most relevant features from OpenFlow network traffic, thereby accelerating processing time and minimizing detection error rates. The experimental results demonstrate that the combination of Random Forest and Gain Ratio achieved a detection accuracy of 99.84% in a controlled testbed environment. This study indicates that the integration of Random Forest and Gain Ratio provides higher accuracy compared to the standard algorithm. The use of Gain Ratio feature selection is proven to optimize both the accuracy and efficiency of the Random Forest algorithm in DDoS detection]]>
Copyrights © 2026
